Skip to Content

Domestic violence continues to be a common and widespread issue throughout the world, far-reaching in its overall impact.

We are a group of students from Victoria University, and we have created this blog post to share our analysis of domestic violence data in Australia from recent years, with a greater focus on NSW using SAP Lumira as a visualization tool.

  • Summarizing the 4-year data that we have obtained from 4 states, NSW, NT, SA, and ACT, here’s the overview of DV occurrence by gender. We can see that the number of DV incidents committed against women is more than twice that of incidents against men:

AU Gender overview.png

Does this pattern hold true across each year and each state?

  • We can see that it is in fact a consistent pattern that women are far more susceptible to DV than men if we break down the data by year and by state:


  • Analyzing the data from the offender aspect, we can see that the majority of DV incidents among women have been committed by their partners, whereas among men, DV is more likely to have been committed by other family members:


     This could be because historically, women have been viewed as “the weaker ***”, and that they have tended to be more emotionally and financially dependent      to their partners than vice versa

  • We then explored the NSW data further to see if we can glean any seasonality in DV occurences over a given year. Looking at the monthly trend from 2010 to 2014, it is evident that domestic violence is at its highest during the months of the December and January:

Seasonality Month.png


     The December numbers could be attributed to the increase in family interaction during the holiday season that potentially leads to higher levels of stress and family conflict. As for January, summer is usually at its hottest during this month. A number of studies have empirically shown that hot temperatures produce increases in aggressive motives and tendencies so hotter years, quarters of years, seasons, and months all yield relatively more aggressive behaviours, such as murders, rapes, assaults, riots and wife beatings, among others.

  • We cross-referenced available daily and hourly data of DV occurences in NSW, and as this graph shows, the majority of DV incidents happen over the weekends. We’d like to call the time frame between 6PM and midnight as the ‘darkest hours’ of the day, as a huge percentage of DV occur within this time period:

Season Day.png

  • We then grouped together the Top 5 and the Bottom 5 Local Government Areas in terms of the number of DV incidents and cross-analyzed them in 4 different demographic aspects:



We can see that the LGAs with the highest DV incidents, represented by the blue bars, have a much higher number of residents whose educational attainment is Year 10 or below and a much higher number of private dwellings without access to internet in comparison to the LGAs with the least DV incidents, represented by the green bars. This suggests that the level of education and information access can be important factors in campaigns and measures to stop DV.

The third and fourth demographic aspects that we considered were the percentage of residents that receive long-term unemployment benefits, and fertility rates. Again, we can see that the LGAs with the most number of DV incidents have a much higher rate of residents who receive financial assistance from the government, and a much higher fertility rate compared to residents of the LGAs with the least number of DV incidents. Having to depend on the government for sustenance and having more children in the family can both increase financial stress, which can trigger aggressive behavior in family members.

  • Interestingly, the top 5 and bottom 5 LGAs are located in rural and urban areas, respectively, which helps explain the differences in the demographic aspects previously discussed. Residents in urban areas enjoy better education, employment rates and higher living standards compared to those in rural areas, and these factors appear to help keep DV occurrences to a minimum.


Remoteness 2.png

To summarise our key findings:


To report this post you need to login first.

Be the first to leave a comment

You must be Logged on to comment or reply to a post.

Leave a Reply